CN110717062B - Music search and vehicle-mounted music playing method, device, equipment and storage medium - Google Patents
Music search and vehicle-mounted music playing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN110717062B CN110717062B CN201810755938.2A CN201810755938A CN110717062B CN 110717062 B CN110717062 B CN 110717062B CN 201810755938 A CN201810755938 A CN 201810755938A CN 110717062 B CN110717062 B CN 110717062B
- Authority
- CN
- China
- Prior art keywords
- music
- user
- pieces
- similarity
- same
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000003068 static effect Effects 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000013136 deep learning model Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 description 16
- 238000010586 diagram Methods 0.000 description 12
- 230000000717 retained effect Effects 0.000 description 11
- 230000002085 persistent effect Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 239000000047 product Substances 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 239000002355 dual-layer Substances 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Abstract
The disclosure provides a music search and vehicle-mounted music playing method, device and equipment and a storage medium. Searching for music matching the user's music search request; performing de-duplication processing on the same music in the searched music; and pushing the music after the duplication removal to the user. Therefore, when the de-duplicated music is played according to the preset sequence, the condition that different versions of the same piece of music are continuously played is avoided, and therefore the music playing experience of a user can be improved. Particularly, for the user in the driving state, the potential safety hazard caused by the unnecessary operation (switching, suspending and other operations) of the user can be avoided.
Description
Technical Field
The disclosure relates to the technical field of internet, and in particular relates to a music search and vehicle-mounted music playing method, device and equipment and a storage medium.
Background
With the rapid development of the internet, people can acquire various resources from the internet. For example, a user may acquire a music resource through the internet.
Currently, intelligent voice interaction services allow users to initiate music searches through voice. For example, in response to a user speaking a voice command of "play a certain ice rain" by a singer, the search engine returns all singers in the music song library with different versions of a certain ice rain, and then plays sequentially. In many cases, however, the user does not wish to repeatedly listen to different versions of ice rain, which requires the user to control the playback by performing additional operations, such as sending a command to "stop playing" or "next song". This undoubtedly reduces the user experience and such additional operations require a high operating cost for the user, especially in driving situations and such unnecessary operations present a certain degree of safety risk.
Disclosure of Invention
It is an object of the present disclosure to provide a music search and vehicle-mounted music playing scheme capable of solving the above-described problems.
According to a first aspect of the present disclosure, there is provided a music search method including: searching for music matching the user's music search request; performing de-duplication processing on the same music in the searched music; and pushing the music after the duplication removal to a user.
Alternatively, the same music is music with the same or similar lyrics and tunes.
Optionally, the step of performing the deduplication processing on the same music in the searched music includes: searching the same music in the searched music; and for finding out a plurality of pieces of music belonging to the same music, only one piece of music is reserved.
Optionally, the step of searching for the same music among the searched music includes: respectively calculating text similarity and audio similarity between any two pieces of music in the searched music; and judging that the two pieces of music with the text similarity exceeding the first preset threshold and the audio similarity exceeding the second preset threshold are the same piece of music.
Optionally, the step of calculating the text similarity between two pieces of music includes: and calculating the similarity between the lyric texts of the two pieces of music as the text similarity.
Optionally, the step of calculating the similarity between the lyrics text of the two pieces of music comprises: respectively encoding lyric texts of different music in the two pieces of music by using a similar hash algorithm; and calculating the Hamming distance of the coding results of the two pieces of music to obtain the similarity.
Optionally, the step of calculating the audio similarity between two pieces of music includes: respectively carrying out vectorization processing on the audio data of two pieces of music to obtain vectorized representation of the audio data of each piece of music; the similarity between the vectorized representations of the two pieces of music is calculated to obtain the audio similarity.
Optionally, the music search method further includes: for the vectorized representation of at least some of the searched music, index information is established for the vectorized representation using a high-dimensional indexing technique to recall audio-similar music from at least some of the music based on the index information.
Optionally, the step of retaining only one of them comprises: and determining the reserved music according to the attribute information of the music and the preference information of the user.
Optionally, the step of retaining only one of them comprises: the music retained in the plurality of pieces of music is determined according to the matching degree of the music and the music search request and/or according to the heat and/or quality of the music.
According to a second aspect of the present disclosure, there is also provided a vehicle-mounted music playing method, including: receiving a music search instruction of a user; searching music matched with the music searching instruction; performing de-duplication processing on the same music in the searched music; and playing the music after the duplicate removal.
According to a third aspect of the present disclosure, there is also provided a music search apparatus including: the searching module is used for searching music matched with the music searching request of the user; the de-duplication module is used for de-duplication processing of the same music in the searched music; and the pushing module is used for pushing the music subjected to the duplication removal to a user.
Alternatively, the same music is music with the same or similar lyrics and tunes.
Optionally, the deduplication module comprises: the searching module is used for searching the same music in the searched music; and a reservation module for reserving only one of the pieces of music for finding out pieces of music belonging to the same piece of music.
Optionally, the search module includes: the first calculation module is used for calculating the text similarity between any two pieces of music in the searched music; the second calculation module is used for calculating the audio similarity between any two pieces of music in the searched music; and the judging module is used for judging that two pieces of music with the text similarity exceeding a first preset threshold value and the audio similarity exceeding a second preset threshold value are the same music.
Optionally, the first calculating module is configured to calculate a similarity between lyrics texts of two pieces of music as the text similarity.
Optionally, the first computing module includes: the encoding module is used for encoding lyric texts of different music in the two pieces of music respectively by using a similar hash algorithm; and the first similarity calculation module is used for calculating the Hamming distance of the coding results of the two pieces of music so as to obtain the similarity.
Optionally, the second computing module includes: the vectorization processing module is used for respectively vectorizing the audio data of the two pieces of music to obtain vectorized representation of the audio data of each piece of music; and the second similarity calculation module is used for calculating the similarity between the vectorized representations of the two pieces of music so as to obtain the audio similarity.
Optionally, the music search apparatus further includes: and the indexing module is used for establishing index information for the vectorized representation result by using a high-dimensional indexing technology aiming at the vectorized representation result of at least part of the searched music so as to recall music similar to the audio from at least part of the music according to the index information.
Optionally, the retaining module determines the retained music according to attribute information of the music and preference information of the user.
Optionally, the retaining module determines the music retained in the plurality of pieces of music according to a matching degree of the music and the music search request and/or according to a heat and/or quality of the music.
According to a fourth aspect of the present disclosure, there is also provided an in-vehicle music playing device including: the receiving module is used for receiving a music search instruction of a user; the searching module is used for searching music matched with the music searching instruction; the de-duplication module is used for de-duplication processing of the same music in the searched music; and the playing module is used for playing the music after the duplication removal.
According to a fifth aspect of the present disclosure, there is also provided a computing device comprising: a processor; and a memory having executable code stored thereon which, when executed by the processor, causes the processor to perform the method as described in the first or second aspect of the present disclosure.
According to a sixth aspect of the present disclosure there is also provided a non-transitory machine-readable storage medium having stored thereon executable code which, when executed by a processor of an electronic device, causes the processor to perform a method as set out in the first or second aspect of the present disclosure.
According to the music playing method and device, the music which is searched out is subjected to the de-duplication processing, so that when the de-duplicated music is played, the situation that different versions of the same piece of music are continuously played is avoided, and therefore the music playing experience of a user can be improved. Particularly, for the user in the driving state, the potential safety hazard caused by the unnecessary operation (switching, suspending and other operations) of the user can be avoided.
Drawings
The foregoing and other objects, features and advantages of the disclosure will be apparent from the following more particular descriptions of exemplary embodiments of the disclosure as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout exemplary embodiments of the disclosure.
Fig. 1 is a schematic flowchart illustrating a music search method according to an embodiment of the present disclosure.
Fig. 2 is a schematic diagram showing the calculation of whether two songs are identical.
Fig. 3 is a schematic block diagram showing the structure of a music search apparatus according to an embodiment of the present disclosure.
Fig. 4 is a schematic diagram showing the structure of a functional module that the deduplication module may have.
Fig. 5 is a schematic block diagram showing the structure of an in-vehicle music playback apparatus according to an embodiment of the present disclosure.
Fig. 6 is a flowchart illustrating the operation of a music search system according to a specific embodiment of the present disclosure.
FIG. 7 shows a schematic block diagram of a structure of a computing device in accordance with an embodiment of the invention.
Description of the embodiments
Preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While the preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
[ scheme overview ]
Under a vehicle-mounted scene, a user (especially a driver) generally only pays attention to whether the music search result is played timely or not when searching for music, and the user does not pay attention to different versions of the same music, and the playing of the same music belonging to different versions can cause a certain trouble to the user, so that the user is prompted to execute unnecessary operations such as pause, switching and the like to skip the playing of the same music of different versions, and the unnecessary operations can bring a certain potential safety hazard. For example, in the case that the music search request of the user only relates to the music name but does not limit the singer, the user only needs to find the song which the user wants to listen to as soon as possible and play, and the different singers do not need to play the different versions of the music which the singer sings for.
Therefore, in order to improve the music playing experience of the user, in particular the music playing experience of the user during driving. The present disclosure proposes that, when performing a music search in response to a music search request of a user, a deduplication process may be performed for the searched music, for example, only one piece may be reserved for a plurality of pieces of music belonging to the same music. Thus, the search result fed back to the user is deduplicated music. When the de-duplicated music is played according to a preset sequence, the condition that different versions of the same piece of music are continuously played is avoided, and therefore the music playing experience of a user can be improved. Particularly, for the user in the driving state, not only the music playing experience of the user can be improved, but also the potential safety hazard caused by the unnecessary operation (switching, suspending and other operations) of the user can be avoided.
Fig. 1 is a schematic flowchart illustrating a music search method according to an embodiment of the present disclosure.
Referring to fig. 1, in step S110, music matching the user' S music search request is searched for.
The user can send out a music search request through voice, and can also send out the music search request through text input. That is, the music search request may be a voice command issued by the user or text information input by the user. In the case where the music search request is a voice command, the voice command may first be recognized as text information using a voice recognition technique, such as an automatic voice recognition technique (Automatic Speech Recognition, ASR), and then a search may be performed based on the recognized text information.
The music searched for may be songs and/or tracks. Wherein, the song refers to music composed of lyrics and tunes, and the tunes refer to music composed of pure tunes without lyrics.
In step S120, the same music among the searched music is subjected to the deduplication process.
The same music mentioned in this disclosure refers to music of the same or similar content. In terms of content, music mainly includes lyrics and tunes, which may be songs composed of lyrics and tunes, or may be tunes composed of tunes only. Thus, the same music may refer to music with the same or similar lyrics and tunes.
The same music can be searched in the searched music, and then for a plurality of pieces of music belonging to the same music, only one piece of music is reserved so as to realize the de-duplication of the same music. The determination and the deduplication mechanism of the same music will be described in detail below, and are not repeated here.
In step S130, the music after the duplication removal is pushed to the user.
Therefore, when the push result is played, the situation that different versions of the same piece of music are continuously played is avoided, and therefore music playing experience of a user can be improved, and particularly for the user in a driving state, potential safety hazards caused by unnecessary operation of the user can be avoided.
Aspects of the disclosure are further described below.
[ definition of identical music ]
As described above, the same music referred to in this disclosure refers to music of the same or similar content, and may be, for example, music of the same or similar lyrics and tunes. In particular, for music with lyrics (such as songs), the same music should be one with the same lyrics as the tune; for music without lyrics (e.g., a song), the same music is a tune-like music.
Example 1: the music is named the same as the song lyrics and the tune, but the singer or player is different, but such different versions of the music may be considered the same music.
Example 2: music is different in name, lyrics and tunes are the same, and the same music is also considered.
Example 3: the singer or player is the same, the tunes are consistent, the original record and its accompaniment are considered as the same music.
Example 4: music is of the same name, but the song and/or lyrics are different, and is not considered to be the same music.
Example 5: different language versions of a piece of music may be considered not identical. For example, the chinese version and the english version are not considered to be songs of the same name because one of the lyrics is chinese and one of the lyrics is english.
[ judgment of identical music ]
As can be seen from the above definition of the same music, the judgment of the same music mainly includes two aspects, the judgment of whether the lyrics are the same or similar and the judgment of whether the tunes are the same or similar.
In the present disclosure, whether the same music is determined by calculating the text similarity and the audio similarity between two pieces of music. Wherein the text similarity may characterize the similarity between text information (e.g., lyrics) of the music, and the audio similarity may characterize the similarity of tunes (i.e., audio features) of the music. Two pieces of music whose text similarity exceeds a first predetermined threshold while audio similarity exceeds a second predetermined threshold may be determined to be the same piece of music. The specific values of the first predetermined threshold and the second predetermined threshold may be set according to practical situations, for example, may be set to 0.8, which is not limited in this disclosure.
1. Text similarity
The text information of the music may include lyrics, a name of the music (e.g., song name), a singer or player of the music, and the like. Since the same music as referred to in this disclosure mainly refers to the same or similar music on the content. Therefore, the text similarity mainly refers to the similarity between lyrics, that is, the similarity between lyric texts of two pieces of music can be calculated, and the obtained similarity can be used as the text similarity. Of course, as an alternative example of the present disclosure, the similarity of music names (e.g., song title similarity) and the similarity of singers or players of music (e.g., singer similarity) may also be considered in calculating text similarity.
An exemplary description is given below of a text similarity calculation process.
(1) Lyric similarity
Since the lyrics of the same music are not exactly the same, they may be mostly similar and the lyrics are longer in length, e.g. 160 bytes in average length. Therefore, if the similarity between lyrics is calculated using a cosine similarity, a Jaccard (Jaccard) similarity, or the like calculation formula, the calculation result may not reflect the similarity between lyrics of two pieces of music more accurately.
Therefore, the disclosure proposes that similar hash algorithm (simhash) can be used to encode lyric texts of different music in two pieces of music, and the similarity of lyrics can be obtained by calculating the Hamming distance of the encoding results of the two pieces of music. The coding principle and the calculation principle of the Hamming distance related to the similar hash algorithm are mature technologies in the field, and are not repeated here.
(2) Similarity of music names
Before similarity of the music names is calculated, normalization pretreatment can be carried out on the music names, special symbols such as English brackets, middle brackets, signature numbers, brackets and the like in the music names can be removed, and the music names are subjected to uppercase, lowercase, simplified, full-angle and half-angle. And then calculate the similarity of song names. Among them, the similarity of music names can be calculated using a similarity calculation formula such as cosine similarity, jaccard (Jaccard) similarity, and the like. The principle of calculating the cosine similarity and the Jaccard (Jaccard) similarity is a mature technology in the art, and will not be described herein.
(2) Similarity of singers or players
After normalizing the character strings corresponding to singers or players of two pieces of music respectively, the two character strings can be segmented with the name of the singer or player as the minimum granularity, and the mutual inclusion degree of the two character strings is calculated and used as the similarity of the singers or players.
As described above, the text similarity referred to in this disclosure mainly refers to lyric similarity. In the case of referring to the similarity of the music name and the similarity of the singer or player at the same time, a preset weight may be set for the three kinds of similarity, respectively, and then the sum of products of the three kinds of similarity under the corresponding weights is taken as the text similarity.
2. Audio similarity
Firstly, the audio data of two pieces of music can be respectively vectorized to obtain vectorized representation of the audio data of each piece of music, and then the similarity between the vectorized representations of the two pieces of music is calculated to obtain the audio similarity.
As an example, for audio data of different music, it may be sampled using the same time interval and then the missing values may be padded using a difference method. Because the time length of different music is different, the number of data points after sampling is different, therefore, each piece of sampled audio data can be normalized, the average value of all audio data is guaranteed to be the same, then each song is reduced to a fixed length by using discrete cosine transform, and each piece of music is guaranteed to be expressed as a real number vector with the same length. To this end, a vectorized representation of the audio data of the music may be obtained.
For real vectors of the same length for any two pieces of music, cosine similarity can be calculated to obtain audio similarity. But since the calculation amount when calculating the similarity of any two songs is too large. Accordingly, the present disclosure proposes that index information may be established for the vectorized representation of at least some of the searched music using a high-dimensional indexing technique, so that music with the same or similar audio is recalled from at least some of the music according to the index information. The implementation of the high-dimensional indexing technique is a mature technology, and is not described in detail herein.
As shown in fig. 2, for two songs, "Qinghai-Tibet plateau-singer Korean-certain" and "Qinghai-Tibet plateau-singer Li-certain", song name similarity, singer similarity, and lyric similarity may be calculated, respectively, and then the total text similarity may be calculated from the calculated song name similarity, singer similarity, and lyric similarity. When calculating text similarity, the lyric similarity occupies a relatively large scale. And, the two hands of songs may be digitized and then the audio similarity calculated. When both text similarity and audio similarity are high, for example, above a first predetermined threshold and a second predetermined threshold, respectively, the two hands of songs may be considered to be the same song.
[ music reservation mechanism ]
For a plurality of pieces of music determined to belong to the same piece of music, only one of them may be reserved. When which one is reserved is specifically selected, the music attribute information and the attribute information of the user can be comprehensively determined. For example, static and dynamic information of music, and static and dynamic information of a user may be considered to decide the music to be finally retained. The static information of the music may be attribute information such as playing times, music Quality (e.g., studio, live edition, SQ (Super Quality)), etc., and the dynamic information of the music may be attribute information such as popularity of the music, popularity of singers, etc., which are determined based on time degradation. The static information of the user may be a long-term interest preference of the user and the dynamic information may be a short-term (e.g., current) interest preference of the user.
That is, the reserved music may be determined according to attribute information of the music and (long and short period) preference information of the user so that the finally reserved music can meet the personalized needs of the user.
As one example of the present disclosure, music retained in pieces of music belonging to the same music may be determined according to the degree of matching of the music with the music search request, and/or according to the heat and/or quality of the music. For example, the music having the highest matching degree with the music search request of the user among the plurality of pieces of music may be reserved, or the music having the highest heat and/or quality among the plurality of pieces of music may be reserved. In addition, music closest to the preference of the user among the plurality of pieces of music may be retained according to the preference information of the user.
For example, after determining whether or not the pieces of music are identical, for pieces of music determined to belong to the same piece of music, it may be decided which piece of music is finally retained based on the following two retention mechanisms.
(1) The search request (e.g., search terms) includes the singer, with the most recent hottest song of the singer being retained.
(2) No singer is included in the search request (e.g., search terms), and the most recent hottest song is retained.
The most recent index is related to static information of songs, such as playing times and music quality (record studio, live edition), dynamic information of songs, such as popularity of songs and popularity of singers, and also related to a long-term short-term interest factor of users, and can determine which track is reserved in the search by establishing a deep learning model.
Thus far, the music search method of the present disclosure is described in detail.
According to the description, in terms of product morphology, the method and the device for searching songs have the same name, the songs with good audio quality are only reserved for the same song to be played to the user, user experience can be improved, and potential safety hazards caused by unnecessary operation of the user are reduced under the scene of a car airport. In the method for identifying the songs with the same name, the accuracy and recall rate are very high, two songs with the same name can be accurately identified, and even if the song names are different, the singers can accurately identify whether the songs with the same name are the songs with the same name or not through deep mining of the lyrics and the song audio. In the selection of the reserved track, the static information and the dynamic information of the song, the static information and the dynamic information of the user are considered, and the final reserved track is determined, so that different user requirements can be met.
[ vehicle-mounted music playing method ]
The method and the device can be applied to the vehicle-mounted scene and are realized as a vehicle-mounted music playing method. As described above, in a vehicle-mounted scenario, a user (especially a driver) generally only care whether the music search result is played timely when searching for music, and for different versions of the same music, the user does not care, and such playing of the same music belonging to different versions may cause a certain trouble to the user, so that the user is prompted to perform unnecessary operations such as pause, switch, etc. to skip playing of the same music of different versions. For example, in the case that the music search request of the user only relates to the music name but does not limit the singer, the user only needs to find the song which the user wants to listen to as soon as possible and play, and the different singers do not need to play the different versions of the music which the singer sings for.
For this reason, the vehicle-mounted music playing method of the present disclosure can receive a music search instruction of a user and search for music matching the music search instruction. Then, the same music in the searched music can be subjected to the de-duplication processing, and the de-duplicated music is played. Therefore, in the scene that the user is in the driving state, the vehicle-mounted music playing scheme based on the disclosure can play the search result after the duplication elimination for the user, so that potential safety hazards caused by unnecessary operations generated by playing the same music with different versions can be avoided. For the definition of the same music and the duplicate removal process, reference may be made to the above related description, and the description thereof will not be repeated here.
[ music search device ]
The music search method of the present disclosure may also be implemented as a music search apparatus.
Fig. 3 is a schematic block diagram showing the structure of a music search apparatus according to an embodiment of the present disclosure. Wherein the functional blocks of the music search apparatus may be implemented by hardware, software, or a combination of hardware and software implementing the principles of the present invention. Those skilled in the art will appreciate that the functional modules depicted in fig. 3 may be combined or divided into sub-modules to implement the principles of the invention described above. Accordingly, the description herein may support any possible combination, or division, or even further definition of the functional modules described herein.
The following briefly describes functional modules that the music search apparatus may have and operations that each functional module may perform, and details related thereto may be referred to the above description in connection with fig. 1 to 2, which are not repeated here.
Referring to fig. 3, the music search apparatus 400 includes a search module 410, a deduplication module 420, and a push module 430.
The search module 410 is used to search for music that matches the user's music search request. The deduplication module 420 is configured to perform deduplication processing on the same music among the searched music. Wherein the same music can be music with the same or similar lyrics and tunes. The pushing module 430 is configured to push the music after the duplication removal to a user.
As shown in fig. 4, the deduplication module 420 may optionally include a lookup module 421 and a reservation module 423. The searching module 421 is configured to search the searched music for the same music. The retaining module 423 is configured to retain only one of the pieces of music that belong to the same piece of music. The retaining module 423 may determine the retained music according to attribute information of the music and preference information of the user. Alternatively, the retaining module 423 may determine the music retained in the plurality of pieces of music according to a matching degree of the music with the music search request and/or according to a heat and/or quality of the music.
Further, the search module 421 may include a first calculation module 4211, a second calculation module 4212, and a determination module 4213. The first calculation module 4211 is configured to calculate a text similarity between any two pieces of music among the searched pieces of music. The second calculating module 4212 is configured to calculate an audio similarity between any two pieces of music among the searched pieces of music. The judging module 4213 is configured to judge that two pieces of music with text similarity exceeding a first predetermined threshold and audio similarity exceeding a second predetermined threshold are identical.
Alternatively, the first calculating module 4211 may be configured to calculate a similarity between lyrics texts of two pieces of music as the text similarity. For example, the first computing module 4211 may include an encoding module and a first similarity computing module (not shown in the figure). The encoding module is used for encoding lyric texts of different music in the two pieces of music respectively by using a similar hash algorithm. The first similarity calculation module is used for calculating Hamming distance of the coding results of the two pieces of music to obtain similarity.
Alternatively, the second computing module 4212 may include a vectorization processing module and a second similarity computing module (not shown in the figure). The vectorization processing module is used for respectively vectorizing the audio data of the two pieces of music to obtain vectorized representation of the audio data of each piece of music. The second similarity calculation module is used for calculating the similarity between the vectorized representations of the two pieces of music to obtain the audio similarity.
As shown in fig. 3, the music search apparatus 400 may further optionally include an index module 440 shown in a dashed box. The indexing module 440 is configured to, for the vectorized representation of at least some of the searched music, build index information for the vectorized representation using a high-dimensional indexing technique to recall audio-similar music from at least some of the music based on the index information.
[ vehicle-mounted music playing device ]
The present disclosure may also be implemented as a vehicle-mounted music playing apparatus.
Fig. 5 is a schematic block diagram showing the structure of an in-vehicle music playback apparatus according to an embodiment of the present disclosure. The functional modules of the vehicle-mounted music playing device can be realized by hardware, software or a combination of hardware and software for realizing the principles of the invention. Those skilled in the art will appreciate that the functional modules depicted in fig. 5 may be combined or divided into sub-modules to implement the principles of the invention described above. Accordingly, the description herein may support any possible combination, or division, or even further definition of the functional modules described herein.
The following briefly describes functional modules that may be provided in the vehicle-mounted music playing device and operations that may be performed by the functional modules, and details related thereto may be referred to the above description of the vehicle-mounted music playing method, which is not repeated herein.
Referring to fig. 5, the vehicle-mounted music playing device 600 includes a receiving module 610, a searching module 620, a deduplication module 630, and a playing module 640.
The receiving module 610 is configured to receive a music search instruction from a user. The search module 620 is used to search for music that matches the music search instruction. The deduplication module 630 is configured to perform deduplication processing on the same music in the searched music. The play module 640 may play the deduplicated music. For functional modules and implementation principles that the deduplication module 630 may have, reference may be made to the description above in connection with fig. 4, and a detailed description is omitted here.
[ application example ]
Fig. 6 is a flowchart illustrating the operation of a music search system according to a specific embodiment of the present disclosure. Wherein circled numbers in the figures show the flow of processing of the data.
Referring to fig. 6, the voice query of the user may be collected by the voice terminal 710 first, and then the voice query may be converted into text query information by the processing module 720, and the search intention of the user may be parsed, for example, the song name and singer information the user desires to search. The parameters entered by the processing module 720 may then be consolidated by the platform 730 and a music search service invoked, such as the consolidated data may be sent to the music search service system 740 for searching.
Specifically, the parsed intent information may be first supplemented by the intent recognition module 741 to augment recall results. For example, personalization and scenerization factors may be considered to supplement search intent. For example, the user searches for "play an passion song", the recognition result of the processing module 720 is "passion song", and directly searching for music does not have an ideal return result. The list of which song ids, album ids, and album ids should be played for "play passion songs" is determined analytically by the intent recognition module 741. The analysis result of the intention recognition module 741 is different for different users and different scenes. That is, the intent recognition module 741 may combine the long-term and short-term music interests of the user, as well as the context in which the user is located, such as: information such as weather, location, destination, time, vehicle condition, etc. are used to supplement the search intent.
The search module 742 may enable a search service using the returned result of the intent recognition module 741 to search the music library for basic information of the corresponding song, such as basic information including song name, singer, number of plays, duration of play, play link, etc.
The ranking module 743 may rank the search results of the search module 742 and deduplicate the same song, and then return the ranking results to the platform 730. The platform 730 sends the results to the processing module 720, and the processing module 720 may instruct the voice terminal 710 to play the search results according to the predetermined arrangement sequence.
[ computing device ]
Fig. 7 illustrates a schematic diagram of a computing device that may be used to implement the data processing of the music search and vehicle-mounted music playing methods described above according to an embodiment of the present disclosure.
Referring to fig. 7, computing device 800 includes memory 810 and processor 820.
Processor 820 may be a multi-core processor or may include multiple processors. In some embodiments, processor 820 may comprise a general-purpose main processor and one or more special coprocessors such as, for example, a Graphics Processor (GPU), a Digital Signal Processor (DSP), etc. In some embodiments, processor 820 may be implemented using custom circuitry, for example, an application specific integrated circuit (ASIC, application Specific Integrated Circuit) or a field programmable gate array (FPGA, field Programmable Gate Arrays).
Memory 810 may include various types of storage units, such as system memory, read Only Memory (ROM), and persistent storage. Where the ROM may store static data or instructions that are required by the processor 820 or other modules of the computer. The persistent storage may be a readable and writable storage. The persistent storage may be a non-volatile memory device that does not lose stored instructions and data even after the computer is powered down. In some embodiments, the persistent storage device employs a mass storage device (e.g., magnetic or optical disk, flash memory) as the persistent storage device. In other embodiments, the persistent storage may be a removable storage device (e.g., diskette, optical drive). The system memory may be a read-write memory device or a volatile read-write memory device, such as dynamic random access memory. The system memory may store instructions and data that are required by some or all of the processors at runtime. Furthermore, memory 810 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), magnetic disks, and/or optical disks may also be employed. In some implementations, memory 810 may include a readable and/or writable removable storage device such as a Compact Disc (CD), a read-only digital versatile disc (e.g., DVD-ROM, dual layer DVD-ROM), a read-only blu-ray disc, an super-density optical disc, a flash memory card (e.g., SD card, min SD card, micro-SD card, etc.), a magnetic floppy disk, and the like. The computer readable storage medium does not contain a carrier wave or an instantaneous electronic signal transmitted by wireless or wired transmission.
The memory 810 has stored thereon executable code that, when executed by the processor 820, causes the processor 820 to perform the music search and vehicle-mounted music playing methods described above.
The music search and on-vehicle music playing method, apparatus and computing device according to the present invention have been described in detail hereinabove with reference to the accompanying drawings.
Furthermore, the method according to the invention may also be implemented as a computer program or computer program product comprising computer program code instructions for performing the steps defined in the above-mentioned method of the invention.
Alternatively, the invention may also be embodied as a non-transitory machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) having stored thereon executable code (or a computer program, or computer instruction code) which, when executed by a processor of an electronic device (or computing device, server, etc.), causes the processor to perform the steps of the above-described method according to the invention.
Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein may be implemented as electronic hardware, computer software, or combinations of both.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems and methods according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of embodiments of the invention has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (13)
1. A music search method, comprising:
searching for music matching the user's music search request;
searching the same music in the searched music;
for searching a plurality of pieces of music belonging to the same music, only one piece of music is reserved; wherein, the step of only preserving one of them includes: determining reserved music by establishing a deep learning model according to attribute information of the music and long-short-period preference information of a user; the attribute information of the music comprises static information and dynamic information of the music, wherein the static information of the music comprises playing times and music quality, and the dynamic information of the music comprises popularity of the music and popularity of singers, which are determined based on time attenuation; and
pushing the music after the duplication removal to the user.
2. The music search method according to claim 1, wherein,
the same music is music with the same or similar lyrics and tunes.
3. The music search method according to claim 1, wherein the step of finding the same music among the searched music comprises:
respectively calculating text similarity and audio similarity between any two pieces of music in the searched music; and
and judging that two pieces of music with the text similarity exceeding a first preset threshold and the audio similarity exceeding a second preset threshold are the same music.
4. A music search method according to claim 3, wherein the step of calculating a text similarity between two pieces of music includes:
and calculating the similarity between the lyric texts of the two pieces of music as the text similarity.
5. The music search method according to claim 4, wherein the step of calculating a similarity between lyrics texts of two pieces of music includes:
respectively encoding lyric texts of different music in the two pieces of music by using a similar hash algorithm; and
and calculating the Hamming distance of the coding results of the two pieces of music to obtain the similarity.
6. A music search method according to claim 3, wherein the step of calculating the audio similarity between two pieces of music includes:
respectively carrying out vectorization processing on the audio data of the two pieces of music to obtain vectorized representation of the audio data of each piece of music;
and calculating the similarity between the vectorized representations of the two pieces of music to obtain the audio similarity.
7. The music search method according to claim 6, characterized by further comprising:
for a vectorized representation of at least some of the searched music, index information is established for the vectorized representation using a high-dimensional indexing technique to recall audio-similar music from the at least some music based on the index information.
8. A vehicle-mounted music playing method, characterized by comprising:
receiving a music search instruction of a user;
searching music matched with the music searching instruction;
searching the same music in the searched music;
for searching a plurality of pieces of music belonging to the same music, only one piece of music is reserved; wherein, the step of only preserving one of them includes: determining reserved music by establishing a deep learning model according to attribute information of the music and long-short-period preference information of a user; the attribute information of the music comprises static information and dynamic information of the music, wherein the static information of the music comprises playing times and music quality, and the dynamic information of the music comprises popularity of the music and popularity of singers, which are determined based on time attenuation; and
playing the music after the duplication removal.
9. A music search apparatus, comprising:
the searching module is used for searching music matched with the music searching request of the user;
the searching module is used for searching the same music in the searched music; and
the reservation module is used for reserving only one of the pieces of music which belong to the same piece of music; wherein, the step of only preserving one of them includes: determining reserved music by establishing a deep learning model according to attribute information of the music and long-short-period preference information of a user; the attribute information of the music comprises static information and dynamic information of the music, wherein the static information of the music comprises playing times and music quality, and the dynamic information of the music comprises popularity of the music and popularity of singers, which are determined based on time attenuation; and
and the pushing module is used for pushing the music subjected to the duplication removal to the user.
10. The music search apparatus according to claim 9, wherein,
the same music is music with the same or similar lyrics and tunes.
11. A vehicle-mounted music playing device, characterized by comprising:
the receiving module is used for receiving a music search instruction of a user;
the searching module is used for searching music matched with the music searching instruction;
the searching module is used for searching the same music in the searched music; and
the reservation module is used for reserving only one of the pieces of music which belong to the same piece of music; wherein, the step of only preserving one of them includes: determining reserved music by establishing a deep learning model according to attribute information of the music and long-short-period preference information of a user; the attribute information of the music comprises static information and dynamic information of the music, wherein the static information of the music comprises playing times and music quality, and the dynamic information of the music comprises popularity of the music and popularity of singers, which are determined based on time attenuation; and
and the playing module is used for playing the heavy music.
12. A computing device, comprising:
a processor; and
a memory having executable code stored thereon, which when executed by the processor causes the processor to perform the method of any of claims 1-8.
13. A non-transitory machine-readable storage medium having stored thereon executable code, which when executed by a processor of an electronic device, causes the processor to perform the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810755938.2A CN110717062B (en) | 2018-07-11 | 2018-07-11 | Music search and vehicle-mounted music playing method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810755938.2A CN110717062B (en) | 2018-07-11 | 2018-07-11 | Music search and vehicle-mounted music playing method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110717062A CN110717062A (en) | 2020-01-21 |
| CN110717062B true CN110717062B (en) | 2024-03-22 |
Family
ID=69208176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810755938.2A Active CN110717062B (en) | 2018-07-11 | 2018-07-11 | Music search and vehicle-mounted music playing method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110717062B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11294954B2 (en) | 2018-01-04 | 2022-04-05 | Audible Magic Corporation | Music cover identification for search, compliance, and licensing |
| US11816151B2 (en) * | 2020-05-15 | 2023-11-14 | Audible Magic Corporation | Music cover identification with lyrics for search, compliance, and licensing |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007164914A (en) * | 2005-12-15 | 2007-06-28 | Matsushita Electric Ind Co Ltd | On-vehicle music reproducing device |
| CN101071422A (en) * | 2006-06-15 | 2007-11-14 | 腾讯科技（深圳）有限公司 | Musicfile search processing system and method |
| CN104484379A (en) * | 2014-12-09 | 2015-04-01 | 百度在线网络技术（北京）有限公司 | Method and device for determining relation among musical entities and inquiry processing method and device |
| CN104715770A (en) * | 2015-03-11 | 2015-06-17 | 广东欧珀移动通信有限公司 | Music playing method and device |
| CN104991943A (en) * | 2015-07-10 | 2015-10-21 | 百度在线网络技术（北京）有限公司 | Music searching method and apparatus |
| CN105138617A (en) * | 2015-08-07 | 2015-12-09 | 中国人民大学 | Music automatic positioning and annotation system and method |
| CN105677690A (en) * | 2014-12-09 | 2016-06-15 | 现代自动车株式会社 | Method and device for generating music playlist |
| CN105930429A (en) * | 2016-04-19 | 2016-09-07 | 乐视控股（北京）有限公司 | Music recommendation method and apparatus |
| CN106919662A (en) * | 2017-02-14 | 2017-07-04 | 复旦大学 | A kind of music recognition methods and system |
| CN107247769A (en) * | 2017-06-05 | 2017-10-13 | 北京智能管家科技有限公司 | Method for ordering song by voice, device, terminal and storage medium |
-
2018
- 2018-07-11 CN CN201810755938.2A patent/CN110717062B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007164914A (en) * | 2005-12-15 | 2007-06-28 | Matsushita Electric Ind Co Ltd | On-vehicle music reproducing device |
| CN101071422A (en) * | 2006-06-15 | 2007-11-14 | 腾讯科技（深圳）有限公司 | Musicfile search processing system and method |
| CN104484379A (en) * | 2014-12-09 | 2015-04-01 | 百度在线网络技术（北京）有限公司 | Method and device for determining relation among musical entities and inquiry processing method and device |
| CN105677690A (en) * | 2014-12-09 | 2016-06-15 | 现代自动车株式会社 | Method and device for generating music playlist |
| CN104715770A (en) * | 2015-03-11 | 2015-06-17 | 广东欧珀移动通信有限公司 | Music playing method and device |
| CN104991943A (en) * | 2015-07-10 | 2015-10-21 | 百度在线网络技术（北京）有限公司 | Music searching method and apparatus |
| CN105138617A (en) * | 2015-08-07 | 2015-12-09 | 中国人民大学 | Music automatic positioning and annotation system and method |
| CN105930429A (en) * | 2016-04-19 | 2016-09-07 | 乐视控股（北京）有限公司 | Music recommendation method and apparatus |
| CN106919662A (en) * | 2017-02-14 | 2017-07-04 | 复旦大学 | A kind of music recognition methods and system |
| CN107247769A (en) * | 2017-06-05 | 2017-10-13 | 北京智能管家科技有限公司 | Method for ordering song by voice, device, terminal and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 范春晓.《Web数据分析关键技术及解决方案》.2017,第66页. * |
| 韩圣龙.《数字音乐信息组织》.2005,第37页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110717062A (en) | 2020-01-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10515133B1 (en) | Systems and methods for automatically suggesting metadata for media content | |
| US7461122B2 (en) | Music delivery system | |
| US20070106405A1 (en) | Method and system to provide reference data for identification of digital content | |
| US20140172910A1 (en) | Music recommendation system and method for vehicle | |
| US10628385B2 (en) | Virtual collection of entities in sync process | |
| KR101942459B1 (en) | Method and system for generating playlist using sound source content and meta information | |
| WO2008004181A2 (en) | Methods and devices for managing a play-list of digital content | |
| US20190147863A1 (en) | Method and apparatus for playing multimedia | |
| US20150334204A1 (en) | Intelligent auto-caching of media | |
| CN110717062B (en) | Music search and vehicle-mounted music playing method, device, equipment and storage medium | |
| CN108766451B (en) | Audio file processing method and device and storage medium | |
| KR101316627B1 (en) | Method and apparatus for recommending music on based automatic analysis by user's purpose | |
| CN111428506B (en) | Entity classification method, entity classification device and electronic equipment | |
| KR101713988B1 (en) | Method and apparatus for providing content sending metadata extracted from content | |
| CN105575400A (en) | Method, terminal, server, and system for obtaining song information | |
| US11410706B2 (en) | Content pushing method for display device, pushing device and display device | |
| CN112182327A (en) | Data processing method, device, equipment and medium | |
| KR20070048484A (en) | Apparatus and method for classification of signal features of music files, and apparatus and method for automatic-making playing list using the same | |
| CN111723289A (en) | Information recommendation method and device | |
| CN103870476A (en) | Retrieval method and device | |
| US6847593B2 (en) | Method for constructing compressed audio file system in a compact disc player | |
| US20110072117A1 (en) | Generating a Synthetic Table of Contents for a Volume by Using Statistical Analysis | |
| US9922118B2 (en) | Creating an audio file sample based upon user preferences | |
| KR20160051983A (en) | Music recommendation system for vehicle and method thereof | |
| CN109165283A (en) | Resource recommendation method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40021952 Country of ref document: HK |
|
| TA01 | Transfer of patent application right |
Effective date of registration: 20201125 Address after: Room 603, 6 / F, Roche Plaza, 788 Cheung Sha Wan Road, Kowloon, China Applicant after: Zebra smart travel network (Hong Kong) Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |